As microbial enzymes which catalyze the oxidation of a hydroxymethyl group to a carboxyl group, there are known alcohol dehydrogenase derived from bacteria of the genus Acetobacter [Agric. Biol. Chem. 42, 2331 (1978)], bacteria of the genus Gluconobacter [Agric. Biol. Chem. 42, 2045 (1978)] or bacteria of the genus Pseudomonas [Biochem. J., 223, 921 (1984), methanol dehydrogenase derived from methanol bacteria (Agric. Biol. Chem. 54, 3123 (1990)] and so on. However, all of them are highly substrate-specific and there is no report ever indicating that they act on substrates other than alcohols such as methanol and ethanol.
Meanwhile, as microbial enzymes which act on sugars, such as D-sorbitol, L-sorbose, etc. (these sugars are hereinafter referred to collectively as sorbose), to catalyze the oxidation of hydroxymethyl to carboxyl, there are known D-sorbitol dehydrogenase [Agric. Biol. Chem. 46, 135 (1982)], sorbose dehydrogenase [Agric. Biol. Chem. 55, 363 (1991) and glucose dehydrogenase [Agri. Biol. Chem. 44, 1505 (1980) derived from bacteria of the genus Gluconobacter. However, all of them are highly substrate-specific and lack versatility. It has also been found that certain bacteria isolated from the soil and named Pseudogluconobacter saccharoketogenes produce 2-keto-L-gulonic acid (hereinafter referred to sometimes as 2KGA) (cf. Japanese Patent Application Kokai S-62-228288 and Kokai S-64-85088) and 2-keto-D-glucaric acid from D-glucose or the like (cf. Japanese Patent Application Kokai S-62-228287) in substantial quantities.
Referring to polysaccharides, dextran, for instance, is a generic name denoting many high molecular glucans predominantly composed of alpha-1.fwdarw.6 bonds as derivatized from sucrose by Leuconostoc mesenteroides and other organisms and a variety of chemical modifications have been attempted. However, by the routine chemical reactions, the object product can hardly be obtained with sufficiently high position selectivity, the formation of many byproducts being inevitable. Since many polysaccharides inclusive of dextran are respectively composed of several high molecular homologs varying in molecular weight, chemical modification is accompanied by side reactions to yield a complicated series of byproducts and it is often true that all of such products cannot be structurally identified (Biotransformations in Preparative Organic Chemistry; H. G. Davis et al, Academic Press).
Particularly, chemical oxidation of dextran with an oxidizing agent such as sodium hypochlorite, sodium hypobromite, chlorine, bromine or iodine has been attempted but the products have not been structurally elucidated for certain or the proposed structures have not been fully validated. Regarding the chemical oxidation of dextran, pertinent disclosure can be found in Patel et al, (Japanese Patent Application, Kokai S -61-233001, European Patent Application, Publication No. 0150085).
As mentioned above, .the oxidation of hydroxymethyl-containing saccharides and other compounds with the aid-of microorganisms has heretofore been seriously restricted to monosaccharides and the like by the substrate-specificity of the microorganisms. Furthermore, as mentioned above, chemical modification of the hemiacetal hydroxyl and hydroxymethyl moieties of saccharides, particularly of oligosaccharides and polysaccharides, are accompanied by many side reactions and unavoidably require a complicated procedure for purification. Therefore, a more selective and efficient oxidation technology has been awaited in earnest. The object of the invention is, therefore, to provide a process for producing saccharic acids, viz. sugar carboxylic acids, comprising oxidizing at least one hydroxymethyl group and/or hemiacetal hydroxyl-associated carbon atom in high yield and with high selectivity by means of a microorganism showing a position specificity to a broad range of substrates for the ultimate goal of producing a number of industrially and socially useful substances.